Chronic fatigue syndrome

In 1990, the Royal Australasian College of Physicians (RACP) published a brief position paper on the investigation and management of chronic fatigue syndrome (CFS) in the RACP magazine, Fellowship Affairs. In 1993, as a result of perceived variations in clinical practice, the then Commonwealth Minister for Health (Senator Graham Richardson) established a CFS Review Committee (comprising Dr David Watson [general physician], Dr Bryce Phillips [general practitioner] and Associate Professor Graeme Stewart [clinical immunologist]) to make recommendations on "diagnostic and management regimens that the medical profession would regard as appropriate for sufferers of CFS". The Review Committee approached the RACP for an up-to-date position, and the College passed the request to the Australasian Society of Clinical Immunology and Allergy (ASCIA). In 1994, a fully revised discussion paper prepared by ASCIA1 was circulated to all specialist physicians in Fellowship Affairs, together with a questionnaire, and the paper and survey results were subsequently made available to the Ministerial Review Committee.

In 1995, as a result of the Review Committee's recommendations, the Commonwealth Department of Health funded the Royal Australian College of General Practitioners to conduct a survey of general practitioners' opinions and practices in relation to CFS. The Ministerial Review Committee also recommended the production of consensus guidelines for distribution to all medical practitioners in Australia. Fortuitously, in October 1995, the National Health and Medical Research Council (NHMRC) published Guidelines for the development and implementation of clinical practice guidelines,2 which provided an ideal framework for this purpose. Consequently, in 1996, a multidisciplinary Working Group (including a Consumer Health Forum representative) was established under the auspices of the RACP to develop and disseminate evidence-based guidelines, following the procedures recommended by the NHMRC. The Commonwealth Department of Health and Family Services provided funding.

Submissions were invited from interested practitioners, consumers and patient support groups. Eighty submissions were received from people with CFS, carers, concerned individuals and CFS Societies. The Consumer Health Forum representative produced two documents: A compilation of submissions made by people with chronic fatigue syndrome and others to the Royal Australasian College of Physicians for the investigation of chronic fatigue and management of chronic fatigue syndrome clinical practice guidelines,3 and A CFS health consumer perspective.4 Quotations for the perspective boxes in these guidelines were drawn from these documents.

The working group prepared draft guidelines that were widely circulated in early 1998. Comments were sought from relevant specialist societies, Royal Colleges, the National Health and Medical Research Council, patient support groups, complementary practitioner associations, and interested individual practitioners and consumers. The draft was also made available on the MJA website <http://www.mja.com.au/public/guides/cfs/cfs1.html>.

The draft guidelines attracted widespread comment both as a result of the initial public consultation and over the four years that they remained available on the MJA website. They were subsequently extensively revised and updated, and underwent a limited second round of public consultation. This final version of the guidelines is the result of revisions carried out in the light of comments received.

The evidence contained within published studies was evaluated according to the process outlined in the NHMRC Guidelines for the development and implementation of clinical practice guidelines (see Box A).2 The quality-of-evidence ratings were modified to provide an integrated system for evaluating diagnostic, epidemiological and pathophysiological studies, as well as treatment trials.

Studies were rated primarily according to the rigour of the research methods used. However, since the interpretation of individual studies is often constrained by selection and other biases, replication across different studies performed in independent research centres was considered a key factor in assessing the reliability of evidence. When the available evidence from several well-conducted studies on a particular topic was conflicting, the quality-of-evidence ranking indicated this uncertainty (Level III-4).

Level IV evidence represents consensus opinions of experts, including working group members, based on clinical experience and limited scientific data. Although such statements may inform current practice, they should be interpreted cautiously, as they may undergo future modification in the light of new evidence.

CFS is diagnosed on clinical grounds. It relies on the presence of characteristic symptoms (see Box B), and the exclusion of alternative medical and psychiatric diagnoses. In individual patients, the symptoms of CFS may overlap with other common syndromes such as fibromyalgia and irritable-bowel syndrome, and the primary diagnosis will depend on which symptoms are the most dominant and disabling. People with CFS often have concurrent depression, and this need not exclude the diagnosis.

As similar symptoms can also occur in a range of other disorders (eg, thyroid disease, anaemia, major depression), the first priority in clinical assessment is to exclude alternative explanations. This can be achieved by careful history-taking, physical examination and a restricted set of laboratory investigations.

Joining a patient support group may be valuable for some people. CFS societies can offer individual and group support, education, and advice about access to social welfare agencies (Box C). Individuals may also benefit from the opportunity to exchange information on how to cope with the many practical day-to-day difficulties that arise for those living with this debilitating condition. However, the quality of advice given can vary and it is therefore useful for the doctor to have ongoing knowledge of the activities and attitudes of local support groups.

C: Support and resources for people with CFS14

CFS = chronic fatigue syndrome; HACC = home and community care.

In 1988, the United States Centers for Disease Control proposed the term "chronic fatigue syndrome" to describe a clinical condition defined by a cluster of constitutional and neuropsychiatric symptoms occurring in a distinctive pattern.23 Current diagnostic criteria (see Box B) describe CFS as a syndrome of physical and mental fatigue, usually of acute onset, which is markedly exacerbated by physical activity. Other common symptoms include headaches, myalgia, arthralgia, and post-exertional malaise; cognitive difficulties, with impaired memory and concentration; unrefreshing sleep; and mood changes.16,24-27 The diagnostic criteria also require that the person must have been ill for more than six months and that the symptom complex is associated with substantial disability.

Delineating CFS as a clinical syndrome has facilitated descriptive clinical research to test the validity of the concept, epidemiological studies to document prevalence and to formulate aetiological hypotheses, laboratory studies to test hypotheses about underlying pathophysiology, and research into a range of treatment strategies.28,29 Although a variety of research definitions have been proposed,6,23,24,27,30-33 the current international consensus criteria for CFS6 have gained wide acceptance in the scientific literature.34,35 In routine clinical practice, a diagnosis of CFS may be appropriate even though the requirement of four out of eight additional symptoms is not formally met (see Box B). Such patients (with "idiopathic chronic fatigue"6) can have comparable levels of disability,27 and may also benefit from the assessment and intervention strategies described in these guidelines.

In primary care, up to two-thirds of people presenting with persistent fatigue have some other identifiable medical or psychiatric disorder that accounts for the symptom,36-41 and careful assessment to exclude these is essential before making a diagnosis of CFS.6

The reported prevalence estimates of CFS differ as a consequence of variations in sampling methods, survey instruments and diagnostic criteria, particularly with regard to duration of illness and the extent to which alternative medical and psychiatric diagnoses were excluded (Box 1.1). Early attempts to record the community prevalence suggested a range of 0.002% to 0.04%.30,55,56 These figures appear to be substantial underestimates as a consequence of limitations in sampling or diagnostic protocols.

The true prevalence of CFS can only be determined in large-scale community studies employing adequate case detection and characterisation techniques. In the US and UK, four studies have provided a more realistic estimate of 0.2% to 0.7% (that is, 200-700 cases per 100 000 people).39,40,57,58 In Japan, the community prevalence has been reported to be 1.5%.59

In primary care settings, estimates of the prevalence of CFS are between 0.5% and 2.5%, depending on the intensity of medical, psychiatric and laboratory evaluation (Box 1.1). Preliminary estimates of the incidence of new cases per year of prolonged fatigue or chronic fatigue in primary care are 3%-5%,40,60,61 whereas the incidence of CFS is about 0.4%.40

1.1: The prevalence of fatigue states

---

Prevalence of prolonged fatigue (PF), chronic fatigue (CF) and chronic fatigue syndrome (CFS) in primary care

Study	PF	CF	CFS
Buchwald et al, 1987, USA51	-	21%	-
Kroenke et al, 1988, USA7	23.8%	-	-
David et al, 1990, UK8	10.5%	-	0.16%
Cathebras et al, 1992, Canada9	13.6%	5.7%	-
Bates 1993, USA52	-	27%	0.3%-1.3%
Katerndahl 1993, USA10	6.9%	-	-
McDonald et al, 1993, UK53	-	112%	2.5%
Walker et al, 1993, USA11	6.7%	-	-
Pawlikowska et al, 1994, UK15	-	18.3%	-
Buchwald et al, 1995, USA39	-	19%	0.1%-0.3%
Hickie et al 1996, Australia13	25%	-	0.3%-1.3%
Wessely et al 1997, UK54	-	11.3%	0.5%-2.6%

CFS predominantly affects young adults, with a peak age of onset between 20 and 40 years.30,40,57,62 In samples of patients from treatment centres, CFS appears to be more common in women (typically in a ratio of 2-3:130), but this may be because women attend all levels of medical care more frequently than men.63 CFS does not preferentially affect individuals from upper socioeconomic groups (contrary to the notion of "yuppie flu").30 Rather, some studies suggest that fatigue syndromes may be more common in people from more socially disadvantaged groups.13,40,62,64 One study has suggested that nurses have a high rate of CFS, indicating that specific occupations may be at risk.65

It is unlikely that common, non-specific viral illnesses trigger the onset of CFS, but specific infections, such as mononucleosis, quite commonly do so. A large controlled study in general practice66 found that people presenting with minor symptomatic infections were no more likely to report chronic fatigue subsequently than those presenting for other reasons. By contrast, a prospective cohort study following individuals with serologically confirmed Epstein-Barr virus infection documented the development of a chronic fatigue state that was independent of psychiatric diagnoses.67 In the Australian context it appears that infections such as Q fever and Ross River virus infection may also trigger CFS.68-71

Fatigue is a central feature of many clinical syndromes (see Box 1.2), including CFS, fibromyalgia, irritable bowel syndrome, major depression, anxiety and somatoform disorders.72-82 These syndromes also share other, non-specific symptoms, including musculoskeletal pain, sleep disturbance, neurocognitive impairment and mood changes.83 Fibromyalgia, in particular, is a closely related syndrome, differing mainly in its relative emphasis on musculoskeletal pain rather than fatigue.73,84-89

1.2: Overlapping diagnoses

Prolonged fatigue states are found in fibromyalgia, irritable bowel syndrome, anxiety and depression, as well as in chronic fatigue syndrome

The number of non-specific medical symptoms reported by people with CFS is strongly correlated with the presence of psychological symptoms.16,90 Up to two-thirds of adults with CFS have either prior or concurrent major depression,36,40,60,74,91-98 as do people with fibromyalgia99 and irritable-bowel syndrome.100,101 By comparison, the lifetime rate of comparable depressive disorders in the general community is 15%-25%.102-106 The high rate of comorbidity is not surprising, as current diagnostic criteria for both CFS and major depression (DSM-IV;107 ICD-10108) include fatigue, sleep disturbance and cognitive impairment, and the presence of mood changes is no longer an exclusion criterion for the diagnosis of CFS.

Perhaps the most difficult diagnostic uncertainty between CFS and psychological illness is in relation to "somatoform" disorders (DSM-IV107). In these disorders, people present with medically inexplicable physical symptoms that are hypothesised to be the result of underlying psychological processes. As the causes of CFS are "unexplained", there is an obvious overlap between the diagnostic criteria for the somatoform disorders and CFS.16,90,109-112

A recent international multicentre study attempted to stratify patients diagnosed with CFS in tertiary referral centres, without prior clinical assumptions.113 The results suggested heterogeneity, with variation between centres, but it was not possible to determine whether the hypothesised subgroups (with "classical CFS" versus "multiple somatic" symptoms) lie on a continuum or represent truly distinct aetiological categories. Nor was it clear whether somatic symptoms were the result of a constitutional vulnerability or were secondary to chronic illness. It was concluded that, although stratification was likely to be important in future research, the basis for allocating subcategories remains controversial.

Whether it will ever be possible to neatly separate a "core condition" of CFS16,33 from other "functional somatic syndromes"114 or to successfully delineate aetiological subcategories of CFS patients remains unclear.18,83 Whatever the case, however, in everyday clinical practice "somatisation" and "somatoform" are unhelpful diagnostic labels which are best avoided in patients with CFS (see Chapter 5).

Though not considered a "cause" of CFS, some patients with chronic fatigue report food intolerances that can exacerbate symptoms.115,116 If food intolerance is suspected on clinical grounds, dietary investigation under the supervision of an appropriately qualified physician and dietitian may be warranted.

Some studies have suggested an overlap between CFS and multiple chemical sensitivity (MCS)82,87,117-124 Gulf War syndrome124,125-128 and "sick building" syndrome.129 The existence of these as valid diagnostic or ontological entities is highly contentious82,130-138 and their consideration is beyond the scope of these clinical practice guidelines.

In the early stages of an illness characterised by prolonged fatigue, spontaneous recovery is common.139 After infectious mononucleosis 41% of patients reported prominent fatigue during the acute illness, of whom 71% had prolonged fatigue one month later, 43% at two months, and 9% at six months.67

By contrast, full recovery in patients with established CFS is less common. In prospective studies, rates of self-reported improvement vary from 11%-64%,140-143 and worsening at 12-18 months was reported in 15%-20%.140,143 A US population surveillance study estimated a cumulative five-year recovery rate of 31%.144

The long-term outcome of CFS has been evaluated mostly in people treated within tertiary referral settings.140,145-150 Such patient samples are biased towards chronic illness and limited patterns of recovery.60,91,139 Patient reports drawn from self-help group populations show similar biases with respect to functional impairment.146 In an Australian study conducted in a specialist setting,150 65 of 103 patients (63%) who had had symptoms for about five years reported abatement of symptoms and improvement in functional capacity over the next three years, but complete recovery was uncommon (6%). During follow-up, patients were very unlikely to develop other medical disorders (2%), but a significant number did develop other psychological disorders (19%), notably major depression and anxiety. Similar outcomes were confirmed in several other retrospective studies from tertiary referral centres.

Factors associated with poorer outcomes include illness duration, subjective cognitive impairment and somatic symptoms,143-151 as well as high levels of fatigue or functional impairment and a low sense of control over symptoms.140 Outcome has not been found to be associated with sex or life stress events,146-152 or with laboratory parameters, such as viral antibody titres and immunological measures (including T-cell-subset measurements).153

At the more severe end of the clinical spectrum, although improvement over time can occur, the prognosis for recovery is poor.154,155 Patients who have had CFS for more than 10 years are more disabled than those with shorter-duration illness, and have significantly more severe symptoms (particularly cognitive impairment) and more frequent symptoms of fibromyalgia.156

Among 2075 people followed up in 19 published studies of the outcome of prolonged fatigue and CFS, there was one death by suicide and two unrelated deaths.139 These studies included mean follow-up periods ranging from six months to four years, suggesting that suicide rates and overall mortality are not increased in people with CFS.

In studies of children and adolescents with CFS the outcome is significantly better than in adults. Two studies evaluating chronic fatigue in children reported that 77%-94% recovered or their condition improved.157,158 The average duration of illness is 2-4 years [see Chapter 4].159-163

The pathophysiological basis of CFS is unclear. The leading hypotheses put forward over the past decade are summarised in Boxes 1.3 to 1.7 and include:

• a unique pattern of infection with a recognised or novel pathogen;164

• altered central nervous system (CNS) function resulting from an abnormal immune response against a common antigen;16,165,166

• a neuroendocrine disturbance;167,168

• a neuropsychiatric disorder with clinical and neurobiological aspects suggesting a link to depressive disorders;169 and

• a psychologically determined response to infection or other stimuli occurring in "vulnerable" individuals.110,170-172

Other hypotheses exist but have not been scientifically evaluated. The probable heterogeneity within patient groups labelled as having CFS18,28,82,83,113 makes it highly likely that there are multiple contributing factors in the disorder.

1.7: Potential central nervous system pathways to chronic fatigue syndrome

The financial impact of CFS on those affected and on the community has been evaluated.22 A conservative Australian estimate of the direct costs (those incurred in diagnosis and management) was $1936 per case per annum (in 1988/89 dollars). After inclusion of indirect costs (from lost productivity associated with the disorder) the aggregate annual cost of CFS was $9436 per case (1988/89 dollars). In 2000/01 dollars, this represents a direct cost of $2764 per case and an annual aggregate cost of $13 471 to the community. Based on a conservative assumption of a community prevalence of CFS of 0.2% (200 cases per 100 000 population), this implies an annual cost to the Australian community of $525 million.

Patients who complain of persisting "fatigue" or "tiredness" may be describing any one of a diverse range of clinical phenomena, ranging from muscle weakness to dyspnoea or depressed mood. The initial task is to clarify the nature of the "fatigue". Fatigue, like pain, is intrinsically a brain-mediated sensation. As with pain, most people report that the fatigue is experienced as a peripheral phenomenon, apparently occurring in musculoskeletal regions. When questioned closely, most people with CFS report that they also experience "mental fatigue", typically precipitated by complex neuropsychological tasks.74,426

To differentiate the various causes of mental and physical fatigue, doctors should focus on the description of the complaint (Box 2.1). Fatigue in people with CFS is typically exacerbated by physical tasks previously achieved with ease, and recovery from periods of worsened fatigue can take hours or even days. Pathological fatigue can be differentiated from:

• somnolence (or "sleepiness"), as it is not relieved by sleep;

• neuromuscular weakness, as people with CFS can generate muscle strength and endurance when circumstances demand;24,400,427 and

• the lack of motivation and loss of pleasure from usual daily activities that characterise depressive illness.

Although these symptoms are common in people with CFS, they are not specific and may occur in a range of other medical and neuropsychiatric disorders. In adults presenting for medical assessment with fatigue states the most common alternative diagnosis to consider is major depression.25,36,39,40,66,74,92-94,96,97 Other commonly detected disorders (Box 2.3) are sleep apnoea, hypothyroidism, anaemia, coeliac disease, chronic hepatitis, panic disorder, generalised anxiety, and somatoform disorders.16,27,39,40,78,81,90,109

When taking a medical history, the questions should focus on key symptoms that might suggest alternative explanations for the fatigue state (see Box 2.1 and Box 2.3). Fatigue accompanied by fever, malaise and weight loss suggests an inflammatory or infective process, and fatigue accompanied by weight gain and cold intolerance may indicate hypothyroidism. Fatigue commonly accompanies many other medical conditions, particularly those directly involving the central nervous system and affecting information processing, the sleep-wake cycle, or arousal mechanisms (eg, multiple sclerosis). Many commonly prescribed medications (such as antihistamines and sedatives) and other substances (such as alcohol, marijuana and amphetamines) cause fatigue directly, or indirectly by disturbing the sleep-wake cycle.

Similarly, physical examination should be directed towards elucidating alternative diagnoses. Apart from minor, non-specific signs of illness, the physical examination in people with CFS is normal.6 Evidence of objective muscle weakness, hard neurological signs, cardiorespiratory disease or fever should alert the doctor to diagnoses other than CFS (see Box 2.3). Although people with CFS often complain of tender cervical lymph nodes, demonstrable lymphadenopathy is not a feature.6

When patients have been definitively diagnosed with a medical or psychiatric condition known to be associated with marked fatigue, a separate diagnosis of CFS is generally not justified.

2.2: Flow chart for the evaluation of persistent fatigue

A formal diagnosis of CFS should not be made without an appropriate psychological evaluation of the patient.6 Although this need not be done by a specialist psychiatrist or psychologist, referral can be useful in selected cases for both diagnostic and treatment purposes. Like the medical evaluation, the psychiatric assessment consists of two distinct parts: the history and the mental state examination.

Brief standardised approaches to psychological evaluation in primary care are available and have been shown to be effective.429 These include self-report questionnaires such as the GHQ-30430 and SPHERE,61,431 or structured interview schedules such as PRIME-MD.432 Important features of the history include prior episodes of anxiety or depression; a past history of multiple, unexplained physical symptoms; and prior alcohol or other substance misuse.

Many people with depressive disorders complain of fatigue or pain, rather than overt psychological symptoms such as tearfulness or sadness. The family history should be reviewed for depressive disorder, self-destructive behaviour or substance misuse. The relationship between the onset of the fatigue state and relevant psychosocial stressors should be noted. Whenever possible, an independent, corroborating history should be sought from a spouse, partner or other family member.

The characteristic mood state of people with CFS is irritation, frustration and transient depression, rather than persistent and profound sadness. This is unlike people with major depression, who report marked anorexia, weight loss, self-reproach and guilt, suicidal plans, persistent loss of motivation or a pervasive loss of pleasure.25,107,108,111

A careful review of the history of ill-health before the onset of CFS is the key to resolving the differential diagnosis of somatoform and somatisation disorders. A long-standing history of frequent medical investigation and treatment for unexplained physical symptoms, persistent fear of medical ill-health despite adequate assessment, preoccupation with unusual physical explanations of illness, and persistent rejection of the potential relevance of psychosocial factors may suggest the diagnosis.107,108

The mental state examination of people with prolonged fatigue should focus on the observed behavioural features rather than simply those reported by the person. These include psychomotor slowing (which may suggest a serious depressive disorder),433,434 demonstrable cognitive impairment (suggesting intoxication, delirium or a dementia syndrome), odd or bizarre interpersonal behaviour (suggesting a psychosis), and hostile, angry or excessively irritable responses (suggesting a personality disorder).

Evaluating a person's risk of suicide is an important task. The major psychological risk factor for suicide is untreated depression. Most people who attempt suicide first present to a healthcare agency, although they typically complain of non-specific symptoms such as poor sleep, poor appetite and tiredness rather than depressed mood.435-437 Other risk factors for suicide include being male, social isolation, concurrent drug and alcohol use and access to lethal means.438,439

The functional impairment of people with CFS has been shown to be similar to or greater than that of people with other chronic disabling medical conditions (eg, multiple sclerosis)440 and psychological conditions (eg, major depression).20,21 Accordingly, the patient's current level of disability should be carefully assessed, with a review of the duration and intensity of physical activity that can be undertaken without precipitating prolonged fatigue. For example, it may be evident that an adolescent's 45-minute walk to school produces fatigue and other symptoms that last all day. At the severe end of the spectrum of CFS, people may be housebound and experience profound fatigue simply from the necessities of self-care, such as showering or dressing.

A diagnosis of CFS is made after six months or more of disabling symptoms. By this time, people with CFS are commonly in crisis with their school or workplace because of the accumulated time lost as a result of the illness. Similarly, by the time of diagnosis, parents, friends and partners of people with CFS are often questioning the nature of the unexplained illness. The effect of the illness upon the patient's key interpersonal relationships,441-443 work or educational activities should be specifically evaluated. This will enable doctors to act as advocates for their patients by providing appropriate information to relevant individuals and institutions.

Despite the wide range of serological, immunological, virological, psychometric and neuroimaging investigations that have been reported in case-control series of people with CFS (see Boxes 1.3-1.6), no specific diagnostic test for the disorder has emerged.6 For any laboratory test to be accepted as having diagnostic validity, it would need to demonstrate both high sensitivity (ie, almost all people with CFS have a positive result - few false negatives) and high specificity (ie, almost all healthy persons, and people with fatigue not due to CFS, have a negative result - few false positives). In fact, as the diagnosis of CFS currently identifies a heterogeneous group of people,16,113 it is unlikely that a single, reliable diagnostic test will emerge.

The only laboratory tests currently recommended for the routine evaluation of people with fatigue states (Box 2.4) are aimed at detecting alternative medical conditions. The diagnostic yield of investigations beyond this restricted list is very low.444-447 If specific alternative diagnoses are suggested by the clinical history or examination (eg, sleep apnoea or multiple sclerosis), further investigations may be warranted.

Many other laboratory procedures have been proposed as "diagnostic tests" by non-medical or alternative practitioners, but have not been subjected to rigorous evaluation. Such "tests" (eg, dark field blood testing for red cell morphology or "candida" identification; stool tests for "dysbiosis"; environmental sensitivity testing) have no basis in evidence and are not recommended.

Once the diagnosis of CFS is made, the doctor should establish an individualised management plan through a process of active discussion with the patient. The available pharmacological and non-pharmacological approaches should first be outlined, along with the role of continuing medical care and the place for physical, social and workplace (or school) rehabilitation programs. The importance of a collaborative approach between patient and doctor should be stressed. The plan should be designed within the framework of the patient's attitudes towards different modes of treatment, bearing in mind the limitations of existing evidence.448

The goal of treatment should be improvement towards and maintenance of maximal achievable functional capacity. While it is very unlikely that any single treatment will provide a "cure", current treatment approaches can result in significant reduction in disability over time.

It is important to give the patient a clear expectation that sustained improvements are rarely achieved in short time frames (days to weeks), but many patients can return to acceptable levels of functioning over longer periods (eg, three to six months).

Sustained improvements are rarely achieved without some setbacks and exacerbations of symptoms along the way. Frequent switching from one form of treatment to another in search of an elusive "cure" should be discouraged, as it is likely to result only in frustration and continuing disability. If patients are made aware of these possibilities at the outset, they will be less likely to abandon useful treatments prematurely.

To facilitate the reduction of disability, active control of key symptoms (eg, pain, sleep disturbance and depressed mood) with standard treatments should be explored. These may include the use of analgesics, non-steroidal anti-inflammatory drugs (NSAIDs), antidepressants or hypnosedatives. If these pharmacological agents prove helpful for the patient, their ongoing use should be reviewed regularly and coordinated with appropriate non-pharmacological forms of care. For example, short-term use of hypnosedatives may assist at the beginning of a structured sleep-wake cycle modification program.449 Similarly, use of analgesics, NSAIDs or both may be necessary during the early phase of a physical rehabilitation approach. Or, if an antidepressant agent (eg, moclobemide) improves a patient's subjective sense of energy and wakefulness,450 this can provide an opportunity to embark on a return to school or work, or a social activity program.

As with other chronic disorders, the patient's attitude to his or her illness experience, understanding of the nature of the disorder and its likely course over time, and the relationship between doctor and patient, are all likely to have a significant impact on long-term outcome.451 Doctors who take an active approach to providing accurate information and to discussing key issues with their patients on an ongoing basis are likely to achieve the best results. This does not mean that the patient and doctor need necessarily agree about all treatment decisions (eg, the use of alternative therapies). It does, however, mean that they should agree on realistic goals for the outcomes of conventional medical treatments. The significant non-specific (placebo) response rate in some controlled treatment trials for people with CFS is likely to be a reflection of these essential components of good clinical practice.452,453

As a general principle of good management of patients with CFS, it can be useful to introduce the concept of self-monitoring of key symptoms and associated disability. This can be achieved through a variety of standardised instruments (eg, SPHERE,431 Brief Disability Questionnaire) and activity, sleep-wake cycle or pain diaries. These allow both the doctor and patient to develop an accurate picture of whether progress is being made with a particular treatment, or whether there has been spontaneous improvement over time. Such monitoring may also alert the doctor to the emergence of a change in key symptoms or disability.

When people with CFS develop significant new symptoms, or experience a marked change in symptoms, they should be carefully reassessed. New symptoms should not automatically be assumed to be part of the CFS symptom complex. Within this context the emergence of depression and other psychiatric complications is particularly relevant.450

3.1: Breaking vicious circles - a rationale for therapy

A range of antiviral, immunoregulatory, antidepressant and metabolic drug regimens have been evaluated in double-blind, placebo-controlled trials in people with CFS. Although limited positive responses have been reported, no agent has consistently shown long-term efficacy in well-designed studies.35

Intravenous immunoglobulin (IVIG): Four double-blind, placebo-controlled trials of therapy with IVIG (based on a hypothesis of disturbed immunity in people with CFS) have been published.264,273,454,455 Two of these trials conducted by one research group in Australia produced conflicting results, with the larger, dose-ranging study showing no significant benefit.273,455 IVIG is not recommended for adults with CFS.

Antidepressants and other CNS agents: Given the high rate of depression, and depression-related symptoms such as fatigue, sleep disturbance, poor concentration and irritability in people with CFS, antidepressant therapies have received considerable attention. To date, there has been no evidence that patients respond in the way that would be expected if CFS were simply misdiagnosed or "masked" major depression. However, certain agents have been found to be beneficial for patients with CFS, particularly those with significant mood or sleep disturbances. Moclobemide (a reversible monoamine oxidase inhibitor) has been evaluated in a large double-blind, placebo-controlled trial.450 Limited evidence of benefit was observed, with an improvement in the subjective sense of vigour and energy that was not associated with any alteration in mood. Similarly, selegiline (a specific monoamine oxidase inhibitor) has been reported to relieve tension and anxiety, and improve vigour and sexual relations.456 All of these agents are somewhat "amphetamine-like" in their actions. While this may assist with key symptoms like fatigue, wakefulness and concentration, they should be used cautiously and closely monitored for side effects such as agitation and insomnia. Their most effective use may be in combination with an active sleep-wake cycle approach.449

While the new serotonergic agents are particularly popular for treating major depression, there has been little evidence of their overall usefulness in patients with CFS. In one of the first large trials, fluoxetine (a selective serotonin reuptake inhibitor [SSRI]) showed no more benefit than placebo.457 However, SSRIs may have a place in patients with concurrent major depression or a strong personal or family vulnerability to anxiety or depression. SSRI therapy needs to be closely monitored for adverse side effects such as nausea, agitation and gastrointestinal disturbances in the early stages of treatment. As SSRIs may disturb sleep-wake architecture during the first few weeks of treatment, patients should also be closely monitored for any exacerbation of the CFS symptom complex. From this perspective, the older tricyclic agents and some of the new antidepressants with more sedative properties (eg, nefazodone) may be more suitable for some patients.

Studies of combination therapy with a low dose tricyclic antidepressant and an NSAID in people with fibromyalgia showed beneficial effects on muscle pain and sleep disturbance, but not on fatigue or mood.458,459

Corticosteroids: Two short-term, placebo-controlled trials of low-dose hydrocortisone therapy in patients with CFS showed a reduction in fatigue and improvement in "wellness", but this was associated with a significant depression of adrenal function.460,461 Given the serious morbidity associated with long-term use,462 corticosteroids cannot be recommended for CFS based on current evidence.

Mineralocorticoids: Although some patients have been found to have postural blood pressure changes consistent with neurally mediated hypotension (Box 1.5), mineralocorticoid therapy has not been found to be beneficial.463,464

Rehabilitation, behavioural and cognitive approaches link the principles of good clinical management with varying degrees of graded physical activity and psychological intervention.153,465,466 They have been evaluated not only in patients with CFS, but in a wide variety of other closely related and often chronically debilitating medical and psychological disorders (eg, chronic pain, chronic daily headache, irritable bowel syndrome and chronic depression). Such treatments are not designed to achieve rapid symptomatic relief. Rather, they are aimed at maximising functional capacity over longer periods. For clinical trial purposes, a specific number of structured treatments are packaged over a set time period. In routine clinical practice, however, the initial treatment strategy is usually individualised according to the patient's level of disability and personal preferences, and subsequently modified according to the response. This is sometimes referred to as "pacing".448

The rationale for these approaches is outlined in Box 3.1.

The behavioural component encourages planned and supervised resumption of appropriate physical and mental tasks. A physical activity program is individually designed to take account of the patient's current level of disability.467,468 After a prolonged period of illness and inactivity, new activities are introduced gradually and, most importantly, are "paced" (ie, scheduled to stop before they produce a significant exacerbation of symptoms). Over time the level of activity attempted is slowly increased at a rate determined by the patient's response.

The cognitive component aims to identify beliefs, attitudes and behaviours that may impair recovery.469-471 Examples include a fear that any increased physical activity will cause harm or prolong illness; a belief that all treatment is futile and that only complete rest will help; a belief that complete withdrawal from work, school and social activities is necessary; a belief that occult chronic infection or chemical exposure has caused permanent injury. The existence of such beliefs is ascertained by exploring the person's causal attributions and his or her understanding of the illness.

As simplistic illness attributions may be associated with poor outcomes,470 people with CFS should be encouraged to adopt the widest possible view of the medical, physical, and psychosocial management strategies to assist in coping with the illness.472,473 The doctor and patient should work cooperatively to improve understanding, attitudes and behaviours that can help maximise long-term function.

In general, trials with more substantial differences between the intervention and the control treatment arms show the greatest benefits.474 Active treatment programs that emphasise strong behavioural components (physical activity, rehabilitation) achieve good short-term results, but studies that incorporate a cognitive component produce more sustained long-term improvements. This may be because patients more readily adopt lifestyle changes that help maintain improved functional capacity beyond the formal treatment period.465,475-477

On balance, current evidence suggests that rehabilitative, behavioural and cognitive approaches should be an integral component of managing people with CFS.478 This contrasts with previous beliefs that prolonged rest and social withdrawal should be advocated.479 By the time patients present with established CFS, many have already experimented with prolonged rest and have found it unhelpful. In some, it may be associated with an exacerbation of sleep-wake difficulties and fatigue. Doctors should ensure that patients are informed of the dangers of prolonged rest480 and the psychological risks of social isolation.

People with CFS experience a range of changes in sleep.77,112,328-331,487-491 The most common features are reduced sleep efficiency, longer sleep onset, increased awakenings during sleep, increased total time in bed, and disturbance of circadian rhythm. There is a growing view that sleep disturbance in patients with CFS may be part of a wider abnormality of sleep-wake cycle function, and that restoration of the normal sleep-wake cycle, with consequent normalisation of circadian rhythm, should be an important goal of therapy.

In patients with CFS, behavioural approaches to sleep-wake cycle disturbance are likely to be more successful than pharmacological approaches, as the latter do not induce normal sleep. Cognitive and educational management approaches should be aimed at promoting an understanding of the role of disordered sleep, and dispelling any irrational fears or inappropriate beliefs about sleep. Relaxation training and stress management may be useful for some patients.

Sleeping for longer does not appear to improve physical or mental functioning in patients with CFS, and excessive periods of daytime sleep or frequent napping serve only to further disrupt circadian rhythm. The aim of sleep management is to establish a regular, normalised sleep-wake pattern:

• try to avoid excessive night-time sleep periods;

• avoid going to bed too early in the evening;

• avoid stimulants during the evening period;

• wake at a regular time in the morning (eg, 7am);

• get out of bed at a regular morning time (eg, by 8am);

• reduce (to less than 30 minutes) or abolish daytime naps; and

• engage in daytime physical and mental activities (within the limits of the individual's functional capacity).

While the recommendations above are generally considered helpful for promoting good sleep in a range of sleep disorders, direct evidence for their benefit in CFS is currently lacking. If a patient with CFS has a concurrent primary sleep disorder (eg, sleep apnoea, restless leg syndrome, narcolepsy), this requires specific intervention.

Given the variable clinical course of CFS, the likelihood of spontaneous improvement and the possibility of non-specific (placebo) responses, properly controlled clinical trials are essential for the evaluation of all proposed new treatments153 (see 3.2). Any claim that a particular treatment can "cure" most people with CFS should be regarded with a high degree of scepticism, not least because patient cohorts in CFS treatment trials are generally heterogeneous16,113 and hence unlikely to respond in a uniform fashion. It should be borne in mind that in some controlled trials over 30% of people with CFS may show improvement in the non-specific (placebo) treatment arm.153,450

In general, evaluation of proposed new treatments for people with CFS requires:

• a plausible scientific rationale for the agent or treatment method to be tested, and preliminary findings showing safety and potential efficacy (phase I data); and

• extension of clinical studies beyond short-term, anecdotal or case-series approaches to randomised controlled trials that evaluate long-term treatment outcomes.490,491

The validity of the results of clinical trials is highly dependent on the quality of study design and analysis.492,493

Critical methodological requirements are:

• use of an internationally accepted case definition;6,32

• random assignment to test or comparison groups;

• adequate sample size;

• use of well-characterised outcome measures494 and standardised self-report instruments for measuring fatigue, mood and other key symptoms;32

• independent, blinded assessments of functional status at onset, completion of treatment, and three to six months later (to ensure durability of the treatment effect); and

• reporting of refusal and drop-out rates, and of the type and frequency of adverse side effects.

Even with well-designed trials, positive results should be independently replicated495 before a new treatment is widely promoted to the general public. The use of essential fatty acids for CFS is a case in point.496 Doctors are encouraged to discuss frankly issues about evidence - including what is known and what is not known - to ensure that patients are able to make informed decisions about treatment.

The prevalence of CFS in adolescents in Australia is not known, although in a prevalence study by Lloyd et al30 20% of those diagnosed with CFS were in the 10-19-years age range. In the United States, prevalence estimates among adolescents reported by Marshall160 depended on the method of case acquisition. These ranged from 116 per 100 000 (telephone interview of CFS-like illness in San Francisco) to 22-26 per 100 000 (school nurses reporting diagnosis, Wichita, Reno) and 3 per 100 000 (physician surveillance: referred cases in four centres).497 All socioeconomic groups are represented and females comprise 71%-74% of the total.160,162,454,498

The onset of the illness in children is typically reported to be abrupt, following a suspected or confirmed viral illness in 85%.498 A seasonal variation with peak onset in autumn and winter has been reported in Australia, suggesting an infectious contribution.498 Epstein-Barr virus (EBV) was the most common infection to be documented serologically at the onset of fatigue symptoms (12% of 290 cases); others were cytomegalovirus, Ross River virus, toxoplasmosis, chickenpox and Mycoplasma. Findings are similar in the US, with most adolescents reporting fatigue following a viral illness, most commonly EBV.160,499

An Australian series of 290 adolescents with CFS from a tertiary referral practice showed a clinical picture highly consistent with that characterised in adults. Prolonged fatigue after physical activity was present in all individuals. Headache, loss of ability to concentrate, the excessive need for sleep, excessive fatigue and myalgia following minor activity, nausea, abdominal pain, sore throat without coryza, and a feeling of disturbed balance were present in 87%, with symptoms rated as severe or moderately severe in over 50%. Feelings of depression and sadness were present in more than 50%. Twenty per cent of individuals met diagnostic criteria for fibromyalgia, a similar figure to that reported in other series.157,158,500-502 The course of the illness also varies. About two-thirds of individuals report continuous symptoms with fluctuating levels of severity, and 15% have a relapsing-and-remitting course.159

Adolescents with CFS often feel miserable, frustrated and angry, particularly after several months of illness. However, young people readily differentiate between feeling miserable meaning "fed-up" from miserable meaning "life is not worth living". Some groups report a higher rate of depression in young people with CFS compared with other chronic illnesses,160 but the sample sizes were small and such series are susceptible to selection bias. When somatic symptoms characteristic of CFS are excluded from the commonly used depression scales, only a small proportion have major depression with anhedonia (7%). The rate is higher in adolescents with more severe CFS, particularly when there is a gradual onset and delay in diagnosis.

Anxiety about returning to the school situation is common in children and adolescents with CFS.503 If extreme, however, other conditions that can mimic or complicate CFS should be considered (eg, depression, eating disorders, school refusal syndromes and, rarely, child abuse).448

Clinical improvement with return to normal functioning is frequently reported in adolescents with CFS, although the longest follow-up studies indicate that a small proportion remain unwell. The average duration of illness is 2-4 years with a range of 1-9 years until well.159-163 Illness pattern varies - about two-thirds have continuous symptoms with fluctuating levels of severity, and 15% have a relapsing-and- remitting course.159

One follow-up study reported that improved functioning rather than return to completely normal health was a relatively common outcome after implementation of a structured management program.162 This program consisted of educational support, graduated exercise, symptom relief, social contact and guidance on planned energy use. After an average of three years (range, 1-6 years) 30% were back to normal, and an additional 20% had mild symptoms following vigorous exercise, but with an otherwise normal activity level. Another 20% were functioning at less than 50% of their previous level of activity, including participation in school or work. Greater morbidity was linked with delay in diagnosis and in receiving assistance.

For young people, CFS often poses special problems that relate to their development.504,505 Some problems are specific to CFS, while others relate to the effect of chronic illness on the emotional and social aspects of adolescent development, examples of which are:

• learning to become autonomous;

• developing a sense of body image;

• understanding and developing relationships;

• making career plans;

• dealing with sexual drives; and

• developing value systems.

Loss of time from school, reduced stamina for writing, and difficulty concentrating (being slower to do things and being only able to concentrate for short periods of time) all contribute to significant educational disruption. Perhaps the most significant effect on schooling, however, relates to the loss of social contacts and access to social learning that plays such a large part in school life.

A management plan should be developed in partnership with the young person and his or her family.448 The individual's illness pattern and severity should be taken into account when designing an individualised program. Although few randomised treatment studies have been performed in children, several strategies have been proposed as helpful.161-163,503,506,507

• Symptom management should include treatment of headache, sleep disturbance, nausea, abdominal pain and dysmenorrhea, and muscle aches and pains.

• Depression and anxiety symptoms should be recognised and treated.

• Activities should be undertaken in a "paced" fashion and planned over a weekly period. The young person with CFS should be encouraged to balance social, physical and intellectual activities, and to make a commitment to undertake segments of each component regularly. This allows the individual to regain some control over their life. A gradual increase in physical activity with school attendance, or a graduated exercise program at home, can be incorporated into an overall weekly plan.

• Maintaining social contact should also be given high priority. This may be through school or via extracurricular activities. Contact with support groups can also be helpful.

• Liaison with the school is essential in order to design an education program involving attendance for particular subjects, or organising some school work by distance education with incorporation of social contact. Early planning and implementation of an educational program is desirable (ie, within 4-6 weeks of onset).

Once clinical improvement occurs, a "tailor-made" program for returning to school can be instituted. Returning to school can be anxiety provoking and stressful for young people with CFS because of remarks made by peers or teachers, as well as the resumption of physical, intellectual and social activities. A loss of confidence in social skills and intellectual ability is commonly reported.

In a survey of adolescents with CFS, general practitioners were considered to be most helpful when they recognised the illness, acknowledged its effects, provided ongoing support, and monitored progress.161,162 Specialist paediatricians were found to be helpful in:

• confirming the diagnosis;

• formulating and coordinating a plan of management;

• providing suggestions for symptom management; and

• providing documentation for and referral to education authorities (to arrange distance education, special consideration and special provision for Year 12 assessments, etc).

In the general population, fatigue states form a continuum in terms of severity and duration, and it is only in those with the most severe and persistent symptoms that a diagnosis of 'CFS' may be appropriate. Although the internationally accepted CFS case definition remains the "gold standard" for diagnosis, it is necessarily arbitrary, having been developed for the purpose of making valid comparisons between research studies carried out in different settings. As such, it creates an artificial boundary within the clinical continuum of fatigue states, giving the false impression that a specific clinical "entity" has thereby been delineated.18,29,46,508

In the absence of a clear understanding of aetiology and pathogenesis, the term CFS should be regarded as a descriptive label only.46,508 Diagnostic boundaries are further blurred by the clinical overlap with other conditions such as fibromyalgia, irritable bowel syndrome, neurasthenia, anxiety and depression, in each of which fatigue can occur as a major symptom.508 In each person with chronic fatigue the doctor must exercise clinical judgement in deciding whether CFS is an appropriate diagnostic label.46

Doctors who display the essential therapeutic characteristics of empathy, acceptance of their patient's suffering, non-judgemental style and a commitment to continued care are more likely to make an appropriate diagnosis515 and to minimise the adverse effects of the illness experience.452,513 Conversely, those who reject the patient's illness experience are likely to promote feelings of alienation and perpetuate ill health.516 A qualitative study of people with CFS found that lack of perceived medical support and understanding was associated with increased seeking of alternative medicine.517 Rejection by family, friends, peers and doctors leads many to experience CFS as a "delegitimising" illness.518 Dismissing a patient's suffering as non-existent or imaginary is anti-therapeutic.28

Broaching the issue of psychological factors in causation should be done with caution and sensitivity, avoiding stereotypic value judgements. The hypothesised role of "somatisation"110 is particularly problematic.111 Outdated and simplistic notions of "psychogenesis", with their implications of "imaginary" illness and "unconscious malingering", leave patients feeling stigmatised, guilty and resentful. Pejorative terms reflecting a false dichotomy between "organic" and "functional" disease519,520 are best avoided.

Unwarranted speculation about psychogenesis, based on the outcome of trials of cognitive behavioural therapy in CFS, should also be avoided. This is only likely to further alienate patients and cause resistance to potentially beneficial management strategies. If an effective therapeutic relationship is to develop, doctors must acknowledge that, despite the current lack of understanding of the underlying cause and mechanisms of chronic fatigue, the symptoms are real and the suffering and associated disability is genuine.448

Support groups have gained prominence in many areas of medicine, including in CFS, and they fill an important gap in areas that have traditionally been poorly catered for within the healthcare system. For individuals and families, they provide an opportunity to share experiences and exchange ideas on coping with practical day-to-day difficulties; they disseminate information on availability and quality of medical and government services, and news of research and treatment advances; and they can offer welcome relief from the sense of isolation that some patients feel. Support groups also have a more general advocacy role, for example in lobbying government agencies to improve funding for patient services, in fund-raising for ancillary services and research, and in promoting community awareness of the plight of sufferers (Box C, page S26).

Not all aspects of CFS support groups are necessarily positive. Inevitably, they tend to attract patients with the greatest functional impairment,146 and this may inadvertently reinforce stereotypes of chronicity, disability and dependency. Moreover, the quality of advice can vary within and between groups, so it is important for practitioners to have ongoing knowledge of the activities and attitudes of local support groups.

Whenever possible, doctors should seek to work cooperatively with support groups. If effective dialogue is to be established and maintained, professionals must be sensitive to the concerns of patient groups, particularly in relation to the inappropriate use of pejorative and stigmatising terms.521 Arrogant and dismissive professional attitudes, amplified by polarised press coverage, can contribute to the alienation of patients from traditional medicine.517,522 Poor communication can also perpetuate misconceptions about aetiology, natural history and treatment rationales, which may themselves contribute to disability.48,110,514,523

Many people with CFS struggle to continue working, despite their chronic illness, for reasons such as self-fulfilment, social identity, or economic necessity. The doctor can provide support by appreciating the specific difficulties experienced by CFS sufferers, and suggesting appropriate coping strategies. Limited energy, cognitive impairment, and memory lapses can impair work effectiveness, placing jobs in jeopardy. Arranging flexibility, prioritising work and compensating for deficits are commonly adopted mitigating strategies.524

Unpredictability resulting from the fluctuating nature of fatigue symptoms525 is a significant problem in conforming to a work routine. Flexibility can often be negotiated in the form of shorter hours or a shorter working week, a variable work schedule with breaks for rest as needed, or discretionary task selection to match variations in capacity.524

Many patients choose to stop working, or are unable to continue, either temporarily or permanently. Practitioners should be supportive in helping patients make the most appropriate choices in relation to their own personal priorities. Those for whom "life is career", and whose struggle to keep working is proving unsuccessful, may become deeply depressed, whereas those who see the maintenance of family and social life as a higher priority may find giving up work a more rational and satisfying choice.524

In verifying a diagnosis of CFS, the current international diagnostic criteria (Box A) should be applied, including documentation of the characteristic symptoms, the lack of abnormalities on physical examination and results of the recommended laboratory investigations. A psychiatric evaluation may be indicated to document any psychological comorbidity.

Forming an opinion about the level of disability is a usual requirement in medicolegal assessment. Since CFS is a subjective illness, initial evaluation relies on a systematic review of the patient's self-reported functional capacity and an assessment of whether this is accurate. Corroborating information may be obtained from a partner or other family member, and from other practitioners with detailed knowledge of the patient.

A doctor acting as an assessor or expert witness may be asked to provide an opinion on causation. Uncertainties regarding the aetiology and pathogenesis of CFS should be acknowledged, and conclusions about the role of infection, chemical exposure or the emotional demands of the workplace should be appropriately tentative unless the clinical evidence is clear-cut and compelling.

Opinions about prognosis should be based on the known natural history, taking account of the duration, clinical course and severity of the individual's illness to date, and his or her progress in response to appropriate symptomatic and behavioural management measures. The notion of "permanent" disability is problematic, as most people with CFS improve gradually, and some eventually recover. In people who have been severely disabled and unable to work for more than five years, the probability of substantial improvement within 10 years is less than 10%-20%. This may be regarded as "permanent disability" for medicolegal purposes.

In the absence of evidence of malingering, speculative judgements about unconscious motivation should be avoided. The psychoanalytic concept of "secondary gain" has been misused in medicolegal settings and does not rest on a solid empirical base.526,527 In evaluating patients with CFS, hypothesised secondary gains should be weighed against manifest secondary losses. The notion of "abnormal illness behaviour" is contentious,528 and the term should not be used as a diagnostic label.